A €20 DNA test shows 80% Basque ancestry and finds a second cousin

A user reports that a €20 MyHeritage DNA test returned 80% Basque ancestry and uncovered a second cousin. The fine print: results depend on who has…

English · Original discussion in Spanish · Published

A €20 DNA test shows 80% Basque ancestry and finds a second cousin
Paying €20 for a DNA test and finding relatives you didn't know existed

Is it worth spending €20 and spitting into a tube so a company can tell you where your great-grandparents came from? The answer from the mass consumption of ancestry kits is a resounding yes, with caveats. A user reports that the MyHeritage report gave him a profile of 80% Basque ancestry, 10% Portuguese-Salamanca border heritage, and the rest split into smaller percentages, and that by cross-referencing that data with FamilySearch genealogy he has rebuilt his family tree back to 1800. What began as skepticism —"at first it seemed like a scam to me"— ended in enthusiasm: "they nailed it 100%".

The star finding is not the percentage, but a second cousin he didn't know existed. That's where these tests stop being a toy and become a real genealogical tool, though with fine print.

How kinship calculation works and why it sometimes fails

The arithmetic of inheritance is unforgiving when it comes to averages: you share 50% of your DNA with your parents, 25% with your grandparents, 6.25% with your grandfather's brother's daughter —the second aunt— and 3% with her daughter, the second cousin. On paper, 3% and around 200 centimorgans shared is the typical signature of a second cousin. The problem is that the average hides an enormous spread, and there the user admits he had to correct himself: "I think it's still a second cousin, it seems that sometimes that DNA segment can be larger".

Meiotic recombination does the rest. Between second cousins, the amount of shared DNA can range from about 50 to over 400 centimorgans, so a result above or below the average doesn't invalidate the relationship. Beyond third or fourth cousins, DNA is no longer useful for confirming a specific link and only provides clues: ancestry, shared match groups or rare mutations from a common ancestor. The practical conclusion is that the test identifies close relatives reliably and distant relatives probabilistically.

Your ancestry percentage depends on who has paid for the kit

Here is the Achilles' heel the user himself admits: "it all depends on how many family members you have registered on that website and have bought the DNA kit". The database is not a census, it's a customers' club. If your family branch has never paid up, the algorithm has nothing to compare against and the result falls short. He found no close relatives; second cousins and distant relatives with 1-2% shared DNA, yes, with uncommon surnames as a clue to identify branches.

The ancestry estimate also has a catch. The 80% Basque returned by MyHeritage gets diluted when the same profile is run through other tools. With Vahaduo and Admixture JS, the Basque Country still comes first, trinc by the French Basque Country and regions near Euskadi —Cantabria, Aragon, southwest France—, while Portugal falls behind Castile and León, and Scotland and Ireland sink to the bottom. The technical criticism coming from the more expert sector is that simulated coordinates are not the same as real ones: with the latter, the profile would be closer to an average Spaniard, specifically a Cantabrian with 3% North African and 7% Italian.

The grandmother's brother who appears in no digital archive

Mormon genealogy has a limit the user ran straight into. FamilySearch allowed him to rebuild his tree back to 1800, but one of his grandmother's brothers —an unknown relative— appears nowhere. The only clue came from a Spanish Civil War search engine: he belonged to a disciplinary battalion. Official archives, both ecclesiastical and state ones, only allow digital searches up to 1900; for anything later, in person.

The remaining route is the Civil Registry, which has existed since 1871. As a relative of a deceased person, you can request a literal birth certificate and, if one exists, a death certificate, through the Spanish Ministry of Justice's e-office. That's the fine print of the test's promise: DNA opens the door, but paperwork gets you through it.

Privacy, health and the genetic data business

The most repeated objection is not genealogical, it's about privacy: "then they have your data, they don't just analyze your genetics". The user's answer is that you can use a fake name and surname so the company can't track you. Half true: DNA is the personal identifier par excellence, and the declared identity doesn't change what your genetic material says about you.

The other commercial hook is health. With the raw test data you can consult reports on other websites, though it's worth remembering that an ancestry report is not a medical diagnosis and any health risk interpretation requires a professional. The €20 kit is no substitute for a geneticist.

The identity trench: how Basque do you have to be to be Basque

The conversation predictably devolved into a purity contest. One user complains that with 20% non-Basque and 2% North African in Eurogenes K36 you're already out of the chosen ones' club. Another jokes that for €20 a customer is told what they want to hear: 50% pre-Indo-European Neolithic and another 50% of Celtic druid or Roman patrician, your choice. The joke has substance: ancestry tests sell identity, and identity is a market with inelastic demand.

The most striking figure is the user's own: "I don't think anyone is more Basque than me, he hasn't posted anything of his". Nobody has shown their real coordinates. In this territory, the one who brags the most usually provides the least data.

At these prices, the DNA test has become the cheapest birthday gift that can change someone's family tree. The second cousin exists. The grandmother's brother, for now, does not.

Also available in: Deutsch

Summary of a discussion on Burbuja.info - Foro de economía, actualidad y política., translated from Spanish and reviewed before publication. Read the full discussion (76 replies).

More summaries

All summaries in English →

-
Back